*** empty log message ***

2 files changed, 138 insertions, 132 deletions

diff --git a/ChangeLog-99b b/ChangeLog-99b
index 90ac89fb8db..37cec7c5eee 100644
--- a/ChangeLog-99b
+++ b/ChangeLog-99b
@@ -1,40 +1,26 @@
-Fri Jun 11 14:47:47 1999  Carlos O'Ryan  <coryan@cs.wustl.edu>
-
-	* bin/auto_compile:
-	  Does not compile all the directories in TAO, because it takes
-	  too much space (and time).
-	  Run several more tests in TAO that are stable.
-	  Improve output
-
-Fri Jun 11 14:40:49 1999  David L. Levine  <levine@cs.wustl.edu>
+Fri Jun 11 15:28:44 1999  Douglas C. Schmidt  <schmidt@tango.cs.wustl.edu>
 
-        * bin/create_ace_build: run the directory mode value through
-          eval, to convert it from octal to decimal, if necessary.
-          Thanks to Russ Noseworthy for providing the patch.
+	* ace/OS.h: Need to add #include <signal.h> before <pthread.h>
+ 	  since pthread.h doesn't include signal.h on SunOS 5.7.  Thanks
+ 	  to Rick Weisner <rick.weisner@east.sun.com> for reporting this.
 
-Fri Jun 11 14:31:14 1999  Carlos O'Ryan  <coryan@cs.wustl.edu>
+	* ace/OS: ACE_OS::memchr() should return a const void * rather
+ 	  than a void *.  Thanks to Rick Weisner
+ 	  <rick.weisner@east.sun.com> for reporting this.
 
-        * ace/Log_Msg.cpp:
-          Use ACE_HAS_THR_C_DEST for TSS "destructors", not
-          ACE_HAS_THR_C_FUNC.
-
-Fri Jun 11 14:13:06 1999  David L. Levine  <levine@cs.wustl.edu>
-
-        * include/makeinclude/platform_sunos5_sunc++.GNU: only set
-          TEMPLATE_REPOSITORY for Sun C++ 4.2.  5.0 doesn't need it.
-          Thanks to Dennis C. De Mars <demars@mminternet.com> for
-          reporting this.
-
-Fri Jun 11 13:05:08 1999  David L. Levine  <levine@cs.wustl.edu>
+        * ace/Dirent.i: Don't call closedir() on a NULL dirp because some
+ 	  platforms (such as SunOS 5.7) will segfault!  Thanks to Tom
+ 	  Arbuckle <arbuckle@uran.informatik.uni-bonn.de> for pointing
+ 	  this out.
 
-        * tests/Reactor_Performance_Test.cpp (main): return error status
-          from run_event_loop () call.  And, added a few ACE_ERROR
-          printouts.  Thanks to Steve for mentioned this.
+Fri Jun 11 14:47:47 1999  Carlos O'Ryan  <coryan@cs.wustl.edu>
 
-        * ace/config-sunos5.5.h: properly reverted all the changes of today.
-          The comments are more helpful this way.
+	* bin/auto_compile: Does not compile all the directories in TAO,
+ 	  because it takes too much space (and time).  Run several more
+ 	  tests in TAO that are stable.  Improve output.
 
-Fri Jun 11 13:00:02 1999  Carlos O'Ryan  <coryan@cs.wustl.edu>
+        * ace/Log_Msg.cpp: Use ACE_HAS_THR_C_DEST for TSS "destructors",
+ 	  not ACE_HAS_THR_C_FUNC.
 
         * bin/ACEutils.pm:
         * bin/Process_Unix.pm:
@@ -42,149 +28,153 @@ Fri Jun 11 13:00:02 1999  Carlos O'Ryan  <coryan@cs.wustl.edu>
           with a timeout, ditto for the ACE::waitforfile() routine.  They
           can be used to write more reliable test drivers.
 
-Fri Jun 11 12:10:00 1999  Chris Gill  <cdgill@cs.wustl.edu>
+Fri Jun 11 14:40:49 1999  David L. Levine  <levine@cs.wustl.edu>
 
-        * tests/RB_Tree_Test.{cpp, h (new)}: factored out class template
-          declaration into a .h file to make AIX C++ happy.
+        * bin/create_ace_build: run the directory mode value through eval,
+ 	  to convert it from octal to decimal, if necessary.  Thanks to
+ 	  Russ Noseworthy for providing the patch.
 
-Fri Jun 11 10:09:51 1999  Douglas C. Schmidt  <schmidt@tango.cs.wustl.edu>
+        * include/makeinclude/platform_sunos5_sunc++.GNU: only set
+ 	  TEMPLATE_REPOSITORY for Sun C++ 4.2.  5.0 doesn't need it.
+  	  Thanks to Dennis C. De Mars <demars@mminternet.com> for
+ 	  reporting this.
 
-        * ace/OS: ACE_OS::memchr() should return a const void * rather
-          than a void *.  Thanks to Rick Weisner
-          <rick.weisner@east.sun.com> for reporting this.
+        * tests/Reactor_Performance_Test.cpp (main): return error status
+ 	  from run_event_loop () call.  And, added a few ACE_ERROR
+ 	  printouts.  Thanks to Steve for mentioned this.
 
-        * ace/Dirent.i: Don't call closedir() on a NULL dirp because some
-          platforms (such as SunOS 5.7) will segfault!  Thanks to Tom
-          Arbuckle <arbuckle@uran.informatik.uni-bonn.de> for pointing
-          this out.
+        * ace/config-sunos5.5.h: properly reverted all the changes of
+ 	  today.  The comments are more helpful this way.
+
+        * include/makeinclude/platform_linux_lxpthread.GNU: changed
+ 	  PLATFORM_AIO_SUPPORT to use := instead of =, so that it only
+ 	  invokes the shell once.  Thanks to Lothar Werzinger
+ 	  <lwerzinger@krones.de> for suggesting this.
 
-Fri Jun 11 07:31:20 1999  David L. Levine  <levine@cs.wustl.edu>
+Fri Jun 11 12:10:00 1999  Chris Gill  <cdgill@cs.wustl.edu>
 
-        * include/makeinclude/platform_linux_lxpthread.GNU:  changed
-          PLATFORM_AIO_SUPPORT to use := instead of =, so that it
-          only invokes the shell once.  Thanks to Lothar Werzinger
-          <lwerzinger@krones.de> for suggesting this.
+        * tests/RB_Tree_Test.{cpp, h (new)}: factored out class template
+          declaration into a .h file to make AIX C++ happy.
 
 Thu Jun 10 23:11:36 1999  Kirthika Parameswaran  <kirthika@cs.wustl.edu>
 
-        * ace/Cache_Map_Manager_T.{h, i, cpp}:
-          Added Iterator implementations to the template parameters to be
-          able to build using the g++ compiler. Also removed unnecessary
-          HASH_KEY and COMPARE_KEYS from the template list.
+        * ace/Cache_Map_Manager_T.{h, i, cpp}: Added Iterator
+ 	  implementations to the template parameters to be able to build
+ 	  using the g++ compiler. Also removed unnecessary HASH_KEY and
+ 	  COMPARE_KEYS from the template list.
 
-        * ace/Hash_Cache_Map_Manager_T.{h, i, cpp}:
-          Made appropriate changes to support changes made to
-          Cache_Map_Manager.
+        * ace/Hash_Cache_Map_Manager_T.{h, i, cpp}: Made appropriate
+ 	  changes to support changes made to Cache_Map_Manager.
 
-        * ace/Caching_Strategies_T.{h, i, cpp}:
-          Added an overloaded clear_cache method which allows the user to
-          specify the total number of entries in the cache. Also added the
-          Caching_Strategy_Utility parameter which decouples the actual
-          purging from the caching strategy.
+        * ace/Caching_Strategies_T.{h, i, cpp}: Added an overloaded
+ 	  clear_cache method which allows the user to specify the total
+ 	  number of entries in the cache. Also added the
+ 	  Caching_Strategy_Utility parameter which decouples the actual
+ 	  purging from the caching strategy.
 
-        * ace/Caching_Strategy_Utility.{h,i,cpp}:
-          Added a Null version of the utility to be used with the Null
-          strategy.
+        * ace/Caching_Strategy_Utility.{h,i,cpp}: Added a Null version of
+ 	  the utility to be used with the Null strategy.
 
-        * ace/Cleanup_Strategies_T.{h,i,cpp}:
-          Added the ACE_UNSED_ARGS macro to ward of g++ warnings.
+        * ace/Cleanup_Strategies_T.{h,i,cpp}: Added the ACE_UNSED_ARGS
+ 	  macro to ward of g++ warnings.
 
-        * tests/Map_Manager_Test.cpp:
-          Added performance tests for the Cache_Map_Manager and the
-          Hash_Cache_Map_Manager.
+        * tests/Map_Manager_Test.cpp: Added performance tests for the
+ 	  Cache_Map_Manager and the Hash_Cache_Map_Manager.
 
 Thu Jun 10 20:29:00 1999  Chris Gill  <cdgill@cs.wustl.edu>
 
-        * ace/RB_Tree.h: Made constructors for iterator base class protected
-          instead of protected, so the the base class can only be constructed
-          by the derived forward and reverse iterator classes.  Thanks to
-          Matt Braun for pointing this out.
+        * ace/RB_Tree.h: Made constructors for iterator base class
+ 	  protected instead of protected, so the the base class can only
+ 	  be constructed by the derived forward and reverse iterator
+ 	  classes.  Thanks to Matt Braun for pointing this out.
 
-        * tests/RB_Tree_Test.cpp: Major overhaul of the test to use a
-          test class template that encapsulates the test details and
-          makes the code much more readable.  Then, the main program merely
-          instantiates a test for each of the four possible combinations of
-          key/item type parameterization over int and const char *, and calls
-          the run_test () method of each test instance.
+        * tests/RB_Tree_Test.cpp: Major overhaul of the test to use a test
+ 	  class template that encapsulates the test details and makes the
+ 	  code much more readable.  Then, the main program merely
+ 	  instantiates a test for each of the four possible combinations
+ 	  of key/item type parameterization over int and const char *, and
+ 	  calls the run_test () method of each test instance.
 
 Thu Jun 10 12:44:22 1999  Steve Huston  <shuston@riverace.com>
 
         * tests/UPIPE_SAP_Test.cpp: Disable for Win32 if !ACE_HAS_WINNT4.
-          The test relies on ACE_SPIPE_Acceptor which requires NT4 to work.
+  	  The test relies on ACE_SPIPE_Acceptor which requires NT4 to
+ 	  work.
 
 Thu Jun 10 12:47:52 1999  David L. Levine  <levine@cs.wustl.edu>
 
-        * ace/config-linux-lxpthreads.h: removed ACE_HAS_AIO_CALLS.
-          It's now enabled in the platform_macros.GNU file:
+        * ace/config-linux-lxpthreads.h: removed ACE_HAS_AIO_CALLS.  It's
+ 	  now enabled in the platform_macros.GNU file:
 
         * include/makeinclude/platform_linux_lxpthread.GNU: added
-          automatic detection of AIO support, based on existence
-          of /usr/lib/librt.so.
+ 	  automatic detection of AIO support, based on existence of
+ 	  /usr/lib/librt.so.
 
-        * include/makeinclude/platform_linux_lxpthread.GNU: add
-          -lrt to LIBS only if $(PLATFORM_AIO_SUPPORT) is equal to
-          -DACE_HAS_AIO_CALLS.
+        * include/makeinclude/platform_linux_lxpthread.GNU: add -lrt to
+ 	  LIBS only if $(PLATFORM_AIO_SUPPORT) is equal to
+ 	  -DACE_HAS_AIO_CALLS.
 
-        Thanks to Jake Hamby <jehamby@anobject.com> for reporting that
-        not all Linux platforms have AIO support.
+        Thanks to Jake Hamby <jehamby@anobject.com> for reporting that not
+ 	all Linux platforms have AIO support.
 
 Thu Jun 10 10:17:08 1999  David L. Levine  <levine@cs.wustl.edu>
 
         * ace/config-sunos5.5.h: with Sun C++ 5.0, #define
-          _RWSTD_NO_EXCEPTIONS 1 if exceptions are not enabled.
-          This avoids a warning about throw expressions being
-          ignored with -noex.  See /opt/SUNWspro_5.0/SC5.0/include/CC/stdcomp.h
-          for more information.
+ 	  _RWSTD_NO_EXCEPTIONS 1 if exceptions are not enabled.  This
+ 	  avoids a warning about throw expressions being ignored with
+ 	  -noex.  See /opt/SUNWspro_5.0/SC5.0/include/CC/stdcomp.h for
+ 	  more information.
 
-        * ace/os-patches/vxworks-5.1.txt,ACE-INSTALL.html: added these notes
-          for using ACE with VxWorks 5.1.
+        * ace/os-patches/vxworks-5.1.txt,ACE-INSTALL.html: added these
+ 	  notes for using ACE with VxWorks 5.1.
 
-        * docs/ACE-guidelines.html:  avoid source code filenames that
-          include "Error".  Avoid #include <math.h>.
+        * docs/ACE-guidelines.html: avoid source code filenames that
+ 	  include "Error".  Avoid #include <math.h>.
 
 Wed Jun  9 21:45:24 1999  Douglas C. Schmidt  <schmidt@tango.cs.wustl.edu>
 
         * tests/Svc_Handler_Test.cpp (main): Revised the test a bit to
-          make it more clear what's going on.
+ 	  make it more clear what's going on.
 
         * ace/Svc_Handler.cpp: Fixed a typo in flush() -- we need to
-          update the temp pointer with temp->cont() *not* entry->cont().
+ 	  update the temp pointer with temp->cont() *not* entry->cont().
 
-        * ace/Svc_Handler.cpp (dump): Fixed a typo -- I wasn't
-          comparing the pointer to 0, I was comparing the
-          ACE_Time_Value!
+        * ace/Svc_Handler.cpp (dump): Fixed a typo -- I wasn't comparing
+ 	  the pointer to 0, I was comparing the ACE_Time_Value!
 
 Wed Jun  9 21:10:00 1999  Chris Gill  <cdgill@cs.wustl.edu>
 
-        * ace/config-psos-diab-ppc.h: added ACE_HAS_NONCONST_SELECT_TIMEVAL
+        * ace/config-psos-diab-ppc.h: added
+ 	  ACE_HAS_NONCONST_SELECT_TIMEVAL
 
-        * tests/RB_Tree_Test.cpp: fixed constness issues revealed by SunC++
-          5.0, added return value assertions for insert calls, added typedefs.
+        * tests/RB_Tree_Test.cpp: fixed constness issues revealed by
+ 	  SunC++ 5.0, added return value assertions for insert calls,
+ 	  added typedefs.
 
 Wed Jun  9 20:40:50 1999  Carlos O'Ryan  <coryan@cs.wustl.edu>
 
         * ace/CDR_Stream.h:
-        * ace/CDR_Stream.cpp:
-          Added new methods to support message reassembly in GIOP 1.1
+        * ace/CDR_Stream.cpp: Added new methods to support message
+ 	  reassembly in GIOP 1.1
 
 Wed Jun  9 15:32:37 1999  Douglas C. Schmidt  <schmidt@mambo.cs.wustl.edu>
 
         * ace/Message_Queue_T: Added a lock() accessor method so that
-          other components can access the lock used by a Message Queue.
+ 	  other components can access the lock used by a Message Queue.
 
         * ace/Svc_Handler.cpp: Added better support for timeout handling
-          for the ACE_Buffered_Svc_Handler.
+ 	  for the ACE_Buffered_Svc_Handler.
 
         * ace/Svc_Handler: Split the buffering portion of ACE_Svc_Handler
-          off into a separate class called ACE_Buffered_Svc_Handler to
-          avoid adding extra space and complexity to the existing
-          ACE_Svc_Handler.  Thanks to Irfan for suggesting this, as well
-          as also suggesting ways to improve the performance by caching
-          the current size of the buffer.
+ 	  off into a separate class called ACE_Buffered_Svc_Handler to
+ 	  avoid adding extra space and complexity to the existing
+ 	  ACE_Svc_Handler.  Thanks to Irfan for suggesting this, as well
+ 	  as also suggesting ways to improve the performance by caching
+ 	  the current size of the buffer.
 
         * ace/OS.h: Added a patch to work around some problems with macros
-          for USYNCH_THREAD.  Thanks to David Levine and Russ Noseworthy
-          for this.
+ 	  for USYNCH_THREAD.  Thanks to David Levine and Russ Noseworthy
+ 	  for this.
 
         * examples/Threads/task_five.cpp (main): Improved the form and
           content of this example a bit.
diff --git a/TAO/ChangeLog-99c b/TAO/ChangeLog-99c
index e78fe60e86a..474d1562fe7 100644
--- a/TAO/ChangeLog-99c
+++ b/TAO/ChangeLog-99c
@@ -1,3 +1,23 @@
+Fri Jun 11 12:27:29 1999  Douglas C. Schmidt  <schmidt@tango.cs.wustl.edu>
+
+	* TAO_IDL/fe/lex.yy.cpp.diff: Updated the diff output to handle
+	  the new flex output from idl.ll.
+
+	* TAO_IDL/fe/idl.ll: Made the following two changes to TAO's IDL
+	  flex file.
+	
+	  Removed the "A" through "Z" characters ranges.  Under EBCDIC
+	  the need to be "A through I", "J through R", and "S through Z"
+	  (ditto for the lower case letters).  Perhaps it should just be
+	  changed to be "[ABCDEFGHIJKLMNOPQRSTUVWXYZ]"...
+
+	  Added support for the trigraph "??=" which is the trigraph for
+	  the "#" sign.  The OpenEdition preprocessor inserts a "pragma" 
+	  statement at the start of the preprocessed output using this
+	  trigraph.
+	  
+	  Thanks to Jim Rogers <jrogers@viasoft.com> for these fixes.
+
 Fri Jun 11 14:48:11 1999  David L. Levine  <levine@cs.wustl.edu>
 
         * performance-tests/Cubit/TAO/IDL_Cubit/run_test.pl,
@@ -6,27 +26,23 @@ Fri Jun 11 14:48:11 1999  David L. Levine  <levine@cs.wustl.edu>
 
 Fri Jun 11 14:33:27 1999  Jeff Parsons  <parsons@cs.wustl.edu>
 
-        * TAO/tests/DynAny_Test/test_dynenum.cpp:
-          Changed test code to match the change in enum labels
-          made in the IDL file a while back.
+        * TAO/tests/DynAny_Test/test_dynenum.cpp: Changed test code to
+ 	  match the change in enum labels made in the IDL file a while
+ 	  back.
 
 Fri Jun 11 14:17:20 1999  Carlos O'Ryan  <coryan@cs.wustl.edu>
 
-        * TAO_IDL/include/be_extern.h:
-          Changed several extern "C" functions to plain extern. They were
-          actually compiled with C++ and was giving warnings under Sun/CC
-          5.0 (and errors in other platforms).
-
-Fri Jun 11 13:01:38 1999  Carlos O'Ryan  <coryan@cs.wustl.edu>
+        * TAO_IDL/include/be_extern.h: Changed several extern "C"
+ 	  functions to plain extern. They were actually compiled with C++
+ 	  and was giving warnings under Sun/CC 5.0 (and errors in other
+ 	  platforms).
 
         * tests/MT_Client/run_test.pl:
-        * examples/POA/Forwarding/run_test.pl:
-          Modified to use the new timeouts, just an example for other test
-          developers.
+        * examples/POA/Forwarding/run_test.pl: Modified to use the new
+ 	  timeouts, just an example for other test developers.
 
-        * orbsvcs/tests/Simple_Naming/run_test.pl:
-          It was missing the ACEutils package. Updated to use the process
-          timeouts.
+        * orbsvcs/tests/Simple_Naming/run_test.pl: It was missing the
+ 	  ACEutils package. Updated to use the process timeouts.
 
 Fri Jun 11 11:42:07 1999  Balachandran Natarajan  <bala@cs.wustl.edu>
 
@@ -147,9 +163,9 @@ Thu Jun 10 21:20:19 1999  Carlos O'Ryan  <coryan@cs.wustl.edu>
 
 Thu Jun 10 17:15:37 1999  Jeff Parsons  <parsons@cs.wustl.edu>
 
-        * tao/Exception.cpp: Added include and modified identifier related
-          to the codeset translator implemented for IBM1047. Thanks to Jim
-          Rogers jrogers@viasoft.com for pointing this out.
+	* tao/Exception.cpp: Added include and modified identifier related
+ 	  to the codeset translator implemented for IBM1047. Thanks to Jim
+ 	  Rogers <jrogers@viasoft.com> for pointing this out.
 
 Thu Jun 10 16:51:02 1999  Pradeep Gore  <pradeep@flamenco.cs.wustl.edu>